US8717888B2 - Optimizations for N-way gateway load balancing in fabric path switching networks - Google Patents

Optimizations for N-way gateway load balancing in fabric path switching networks Download PDF

Info

Publication number
US8717888B2
US8717888B2 US13/331,626 US201113331626A US8717888B2 US 8717888 B2 US8717888 B2 US 8717888B2 US 201113331626 A US201113331626 A US 201113331626A US 8717888 B2 US8717888 B2 US 8717888B2
Authority
US
United States
Prior art keywords
switch
group
emulated
gateway
fhrp
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related, expires
Application number
US13/331,626
Other languages
English (en)
Other versions
US20130094357A1 (en
Inventor
Ramkumar Sankar
Naveen Pethandlahally Narayan Gowda
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Cisco Technology Inc
Original Assignee
Cisco Technology Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Cisco Technology Inc filed Critical Cisco Technology Inc
Assigned to CISCO TECHNOLOGY, INC. reassignment CISCO TECHNOLOGY, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: GOWDA, NAVEEN PETHANDLAHALLY NARAYAN, SANKAR, RAMKUMAR
Priority to PCT/US2012/060792 priority Critical patent/WO2013059440A1/en
Priority to EP12783754.0A priority patent/EP2769515B1/de
Priority to CN201280051108.2A priority patent/CN103891216B/zh
Publication of US20130094357A1 publication Critical patent/US20130094357A1/en
Application granted granted Critical
Publication of US8717888B2 publication Critical patent/US8717888B2/en
Expired - Fee Related legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/58—Association of routers
    • H04L45/586—Association of routers of virtual routers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/70—Routing based on monitoring results
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/12—Avoiding congestion; Recovering from congestion
    • H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/35—Switches specially adapted for specific applications
    • H04L49/354—Switches specially adapted for specific applications for supporting virtual local area networks [VLAN]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/35—Switches specially adapted for specific applications
    • H04L49/356—Switches specially adapted for specific applications for storage area networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/70—Virtual switches

Definitions

  • L2MP Layer 2 multipath network
  • FSS fabric path switching System
  • OSPF Open Shortest Path First
  • a data center Ethernet may include a collection of gateway switches at the L2/L3 boundary used to forward frames towards (and receive frames from) destinations outside of the data center.
  • Such gateways may be configured to provide redundancy using a First Hop Routing Protocol (FHRP), such as Hot Standby Router Protocol (HSRP) or Virtual Router Redundancy Protocol (VRRP), and static load balancing may be achieved among the gateways using multiple-group FHRP.
  • FHRP First Hop Routing Protocol
  • HSRP Hot Standby Router Protocol
  • VRRP Virtual Router Redundancy Protocol
  • static load balancing may be achieved among the gateways using multiple-group FHRP.
  • a large data center network may include 16 or even 32 such gateways.
  • HSRP or VRRP does not scale or provide for dynamic load balancing at the L2/L3 boundary as gateways are added (or removed) from an FHRP group. This occurs because the number of paths that must be configured on each access switch grows geometrically with each additional gateway and FHRP group. In such cases,
  • FIG. 1 illustrates an example network computing infrastructure connecting classical Ethernet hosts to a Layer 2 multipath access switch, according to certain embodiments of the present disclosure.
  • FIG. 2 is a block diagram illustrating an example of a L2MP switching device, according to certain embodiments of the present disclosure.
  • FIG. 3 illustrates an example of configuring a MAC table on the L2MP access switch first shown in FIG. 1 , according to certain embodiments of the present disclosure.
  • FIG. 4 illustrates an example of configuring a switch table on the L2MP access switch first shown in FIG. 1 , according to certain embodiments of the present disclosure.
  • FIG. 5 illustrates an example of the switch table of the L2MP access switch, after being updated in response to a gateway switch failure, according to certain embodiments of the present disclosure.
  • FIG. 6 illustrates a method for populating a MAC table on a L2MP access switch, according to certain embodiments of the present disclosure.
  • FIG. 7 illustrates a method for populating a switch table on a L2MP access switch, according to certain embodiments of the present disclosure.
  • FIG. 8 illustrates a method for n-way gateway load balancing in fabric path switching networks, according to certain embodiments of the present disclosure.
  • One embodiment provides an access switch configured to provide n-way load balancing for layer 2 traffic forwarded to multiple gateways in fabric path switching networks.
  • One embodiment includes a switching device which has at least a processor and a memory, wherein the switching device is configured to perform an operation for performing n-way gateway load balancing in a fabric path switching network.
  • the operation itself may generally include the operation comprising receiving, at an access switch, a first network frame from an active gateway router in a first hop routing protocol (FHRP) group of gateway routers.
  • the first frame may include an emulated switch ID assigned to the FHRP group of routers.
  • the access switch adds an entry to a media access control (MAC) table mapping a MAC address associated with the emulated switch to the emulated switch ID.
  • MAC media access control
  • the operation may further include receiving, at the access switch, one or more link state messages from one or more respective standby gateway routers in the FHRP group, Each of the one or more second link state messages includes the emulated switch ID.
  • the access switch adds an entry to a switch table a link on the access switch over which the emulated switch may be reached.
  • the operation may further include receiving, from a host attached to the access switch, a frame with a destination address of the emulated switch.
  • the access switch forwards the frame received from the host to a selected one of the group of gateway routers. That is, the gateway router is selected to load balance frames forwarded to the emulated switch across the group of gateway routers.
  • one FHRP group could be used to designate a first gateway router as being “active” for the first emulated switch.
  • another FHRP group could be configured on the same two gateway routers, where the first gateway router is configured as being “standby” and the second gateway router as being “active” for the second emulated switch.
  • This configuration allows each gateway router to act as a failover for the other.
  • each emulated switch is “active” for at least some traffic (e.g., traffic for different VLANs)
  • neither of the underlying gateway routers remain idle and unused. That is, as each switch forwards traffic there is generally no logical difference between an “active” router and a “standby” router except for ARPs/VIP-pings.
  • Embodiments described herein allow the FHRPs to more fully operate in an active-active model in order to take full advantage of the Layer 2 multi-pathing available when using multiple gateway routers and to extend load balancing to all Layer 3 traffic.
  • more than 2 gateway routers may be present in a large L2MP network and each should be available to forward traffic for any VLAN in the data center.
  • access switches which use the emulated switch as a “first hop” for sending (and receiving) Layer 3 traffic may intelligently snoop link-state messages (e.g., IS-IS or OSPF messages) sent from the gateway routers to learn which links on the access switch may be used to forward traffic to the emulated switch. Doing so allows the access switch to load balance traffic to the identified gateways (i.e., to the emulated switch) as well as dynamically determine which switch links the emulated switch may be accessed over.
  • link-state messages e.g., IS-IS or OSPF messages
  • embodiments presented herein allow inter-working of L2MP and classical Ethernet networks, as well as provide classical Ethernet network with first-hop redundancy.
  • FIG. 1 illustrates the example described above. More specifically, FIG. 1 illustrates a network computing infrastructure 100 connecting classical Ethernet (CE) hosts 125 1-2 to a L2MP access switch 105 , according to certain embodiments of the present disclosure. As shown, the access switch 105 is configured to forward layer 3 (L3) traffic for hosts in a first VLAN (e.g., for host 125 1 ) to a first emulated switch (ES1) 110 and forward L3 traffic for hosts in a second VLAN (e.g., for host 125 2 ) to a second emulated switch (ES2) 112 .
  • L3 layer 3
  • Both emulated switches ES1 110 and ES2 112 are configured using a first gateway router 115 1 and a second gateway router 115 2 .
  • Each gateway router 115 provides a respective uplink (labeled U 1 and U 2 ) to a layer 3 network 120 .
  • the first emulated switch (ES1) includes two gateway routers that are part of an FHRP group, where gateway router 115 1 is active and where gateway router 115 2 provides a standby failover.
  • an FHRP group has been defined where gateway router 115 2 is active and gateway router 115 1 provides a standby failover.
  • FIG. 2 is a block diagram illustrating an example of a L2MP switching device 200 , according to certain embodiments of the present disclosure.
  • the switching device 200 includes ports 205 , a processor/firmware 210 , routing/switching logic 215 , management interface 220 , and memory 225 .
  • the memory 225 includes a MAC table (MTB) 230 , a switch table (STB) 235 , and configuration settings 240 .
  • MTB MAC table
  • STB switch table
  • configuration settings 240 In one embodiment, users may specify configuration settings 240 for the switching device using the management interface 220 .
  • the management interface 220 may be used to enable/disable L2MP routing as well as whether to enable/disable n-way gateway load balancing in the fabric path where a set of routing gateways are used to provide multiple layer 2 outgoing paths for hosts connected to the switching device 200 .
  • the depiction of a switching device 200 shown in FIG. 2 is simplified to highlight aspects of the embodiments described herein, and that, in practice, network switching/routing devices may be configured with a variety of additional functions, features and components.
  • Switching device 200 generally represents a layer 2 switch which provides Ethernet connectivity between a large number of hosts and a set of uplinks to gateway devices.
  • ports 205 include L2MP links 201 (i.e., links to gateway routers) as well as classical Ethernet (CE) links 202 .
  • the L2MP links 201 represent the physical cabling connecting the switching device 200 to a set of gateway routers and the CE links 202 represent the physical cabling connecting the switching device 200 to a collection of hosts.
  • switching device 200 generally corresponds to the access switch 105 of FIG. 1 .
  • the L2MP links 202 connect switching device 200 to other switches within a network fabric (e.g., to the gateway router 115 ), and CE links 202 connect switching device 200 to a set of CE hosts (e.g., hosts 125 ).
  • the access switch may learn an association between a MAC address and a port 205 by inspecting the source address field of an Ethernet frame sent to the switching device 200 by a host connected to one of the CE links 202 .
  • MAC frames sent by a gateway router may include an actual MAC address for that router or include a virtual MAC address when the gateway router is configured as part of an emulated switch.
  • entries in the MTB 230 may provide a mapping between a MAC address identified in the outer source address of the L2MP frame and a switching entity in a fabric path which owns that MAC address.
  • an entry in the MTB 230 may provide a mapping between a virtual MAC address associated with an emulated switch and a switch ID for that emulated switch.
  • the MTB 230 When a frame (whether a CE frame or an L2MP frame) is received listing the virtual MAC address as a destination, the MTB 230 is then used to determine that such a frame should be forwarded to the switching entity identified in the MTB 230 (e.g., that such a frame should be forwarded to the emulated switch). Further, the links over which the identified switching entity may be reached may be identified in the switch table (STB) 235 . More specifically, the STB 235 may include a mapping between one or more switches in the fabric path and the ports 205 over which the switching entity may be reached. In one embodiment, when a switching entity is reachable over multiple links, the switching device 200 may be configured to load balance traffic over the available links. Such load balancing may distribute the traffic relatively equally, but could also favor one link over another (e.g., in the case where one link has a known greater capacity or bandwidth).
  • the access switch may be configured to learn this information by intelligently monitoring FHRP group messages (typically sent to a multicast destination which includes each switch assigned to an FHRP group) and link state messages used to determine loop free routing topologies (including routing topologies with multiple paths between two layer 2 entities).
  • FHRP group messages typically sent to a multicast destination which includes each switch assigned to an FHRP group
  • link state messages used to determine loop free routing topologies (including routing topologies with multiple paths between two layer 2 entities).
  • FIG. 3 illustrates an example of configuring a MAC table on the L2MP access switch 105 first shown in FIG. 1 , according to certain embodiments of the present disclosure.
  • an FHRP group 110 is configured on the gateway routers 115 1-2 , designated to be a first hop gateway for the access switch 105 .
  • gateway router 115 1 is designated as “active” for the FHRP group 110 and gateway router 115 2 is designated as “standby” for the FHRP group 110 .
  • an emulated switch-id is allocated to each gateway router 115 in the FHRP group 110 .
  • access switch 105 learns the entry 305 “VMAC, ES1” in the MTB table from an FHRP hello packet 310 sent from the active gateway router 115 1 to the access switch 105 .
  • the FHRP hello packet 310 is sent as a “MAC-in-MAC” packet with outer-MAC source address referencing the emulated switch ID (i.e., referencing “ES1” in this example).
  • the outer destination address of “MC” or multicast results in the access switch 105 forwarding the hello packet 310 to one or more standby hosts in the FHRP group 110 .
  • the access switch 105 forwards the FHRP hello packet 310 to the gateway router 115 2 (acting as a standby for the FHRP group 110 ). More generally, due to the periodic hello packets, edge switches learn the reachability of “virtual MACs” to be reachable via an “emulated switch-id” assigned to the router sending the hello packet.
  • host A 125 when host A 125 first comes up, it may send an ARP request to learn the MAC address corresponding to an IP address of the default gateway. In one embodiment, only the “active” router for the FHRP 110 group responds to an ARP request. Further, such a response may identify the virtual MAC address associated with the emulated switch (ES1). Doing so allows the CE hosts attached to access switch 105 (e.g., host 125 ) to learn the VMAC address to use as the destination address for frames forwarded to the default gateway.
  • ES1 emulated switch
  • FIG. 4 illustrates an example of configuring a switch table (STB) on the L2MP access switch 105 first shown in FIG. 1 , according to certain embodiments of the present disclosure.
  • gateway routers assigned to an FHRP group are each assigned an emulated switch ID and the access switch learns the emulated switch ID from hello packets sent by the active gateway router in an FHRP group.
  • the access switch 105 may be configured to intelligently monitor link state protocol messages to learn what links the emulated switch ID (learned through FHRP hello messages) is reachable over.
  • gateway routers 115 1-2 may be configured to use a link state protocol (e.g., IS-IS or OSPF) to advertise the reachability of the emulated switch-id (ES1) from each of the routing gateways that are part of the emulated switch, i.e., the routing gateways that are part of the FHRP group.
  • link state protocol e.g., IS-IS or OSPF
  • gateway router 115 1 sends an IS-IS message 405 1 and access switch 105 learns that the emulated switch-id (ES1) may be reached over link 1 (L1).
  • gateway router 115 2 sends an IS-IS message 405 2 and access switch 105 learns that emulated switch (ES1) may also be reached over link 2 (L2).
  • switch table includes an entry 410 “ES1, L1, L2,” indicating that the emulated switch is reachable over both link 1 and link 2.
  • frames with a destination address of “VMAC” may be load balanced over link 1 (L1) and link 2 (L2). More generally, load balancing can be achieved at L2/L3 boundary for as many gateways as present in a data center Ethernet.
  • each gateway participating in one (or more) FHRPs or configured with SVIs listens to FHRP hello from other gateways. Based on the received hellos messages on the VLAN/group, and based on the local configuration, a gateway may determine whether a mismatch exists. That is, one of the gateway routers 115 may determine whether it may receive traffic that cannot be forwarded to its destination. In case of a mismatch, the routing gateway may withdraw from emulated switch membership corresponding to the affected SVIs or FHRP groups. Once the inconsistency is cleared, the gateway router may enroll back into the emulated switch membership. Note that the “withdrawal” and “enrolling” into emulated switch membership may be achieved using the already existing ISIS protocol used in L2MP for advertising the switch-id reachability.
  • the local switch/router when the local switch/router receives a HSRP hello packet with a FHRP group number that is not configured in the local system, that switch/router may conclude that the routers/switches have been configured with group number which is not present in the local system. This might black hole the traffic belonging to that particular group. Accordingly, at this stage the switch is removed from the emulated switch.
  • One or more tracking or status objects can be associated with an FHRP group. Whenever the track object goes down, we decrement the group's priority with the decrement priority associated with the track object. When the priority falls below configured lower threshold the local switch is withdrawn from the emulated switch, concluding that the switch will be unable to forward the data traffic through the upper link. When the tracked object comes back up, then the priority will increase, and once the priority crosses upper threshold, then the switch is added back to emulated switch.
  • FIG. 5 illustrates an example of the switch table of the L2MP access switch 105 , after being updated in response to a gateway switch failure, according to certain embodiments of the present disclosure.
  • an uplink to L3 network 120 on gateway router 115 2 has gone down.
  • gateway router 115 2 determines that it needs to withdraw from the emulated switch ES1 and does so by sending an IS-IS message 515 announcing its withdrawal.
  • access switch 105 modifies an entry 520 in the switch table (STB) to remove the indication that the emulated switch ES1 is reachable over link 2 (L2).
  • the access switch 105 when the access switch 105 receives frames addressed to the virtual MAC address of the emulated switch, it no longer load balances such frames over both link 1 (L1) and link 2 (L2). For example, when a CE host (e.g., host 125 ) forwards a MAC frame 505 to the access switch 105 with a destination address of VMAC, the access switch 105 wraps the frame 505 with a outer MAC header 510 indicating the source address of the access switch 105 “AC 1 ” and a destination address of the emulated switch (i.e., of ES1). However, as the only link to the emulated switch in the STB is L1, the only link used to send frames to the emulated switch is L1.
  • a CE host e.g., host 125
  • the access switch 105 wraps the frame 505 with a outer MAC header 510 indicating the source address of the access switch 105 “AC 1 ” and a destination address of the emulated switch (i.e., of ES1).
  • gateway router 115 2 may re-enroll in the emulated switch, resulting in access switch 105 restoring the entry in the STB indicating the reachability of the emulated switch over link 2 (L2).
  • FIG. 6 illustrates a method 600 for populating a MAC table on a L2MP access switch, according to certain embodiments of the present disclosure.
  • the method 600 begins at step 605 , where the access switch or edge switch within a data center Ethernet network receives an FHRP hello message from a gateway switch participating in an FHRP group.
  • the hello message may be configured according to the known Hot Standby Router Protocol (HSRP) or according to the Virtual Router Redundancy Protocol (VRRP).
  • HSRP Hot Standby Router Protocol
  • VRRP Virtual Router Redundancy Protocol
  • embodiments may be adapted for use with other FHRP protocols as they become available.
  • a given gateway router may participate in multiple FHRP groups (and emulated switches), only the gateway router configured as the “active” one for a given FHRP group sends the hello message.
  • the access switch identifies an emulated switch ID assigned to an emulated switch configured form the gateway routers participating in the FHRP group.
  • the emulated switch ID may be determined from an outer source address field of a MAC-in-MAC packet used send the hello message.
  • the access switch adds the emulated switch ID to the MAC table (MTB). That is, the access switch creates an entry in the MAC table mapping the virtual MAC address to the emulated switch ID.
  • the access switch forwards the hello message to a multicast destination address identified in the outer destination address filed of the outer MAC-in-MAC header. That is, the access switch forwards the hello message according to the FHRP protocol, as would occur without the extra processing to create entries in MTB table.
  • FIG. 7 illustrates a method 700 for populating a switch table on a L2MP access switch, according to certain embodiments of the present disclosure.
  • the method 700 begins at step 705 , where an access switch receives a link state protocol message from a gateway router.
  • the link state protocol message may be configured according to the IS-IS or OSPF link state protocols. However, embodiments may be adapted for use with other link state protocols as they become available.
  • the access switch adds an entry to the switch table indicating the reachability of the emulated switch over one of the links on the access switch. Alternatively, if the IS-IS message indicates that a gateway router is withdrawing from the emulated switch, then the switch table may be updated to remove the appropriate link as providing reachability to the emulated switch.
  • the access switch forwards the link state message according to the link state protocol being used and the content of the particular message.
  • steps 705 , 710 , 715 may be performed for each gateway router included in a given FHRP group. And further, steps 705 , 710 , 715 may be performed dynamically as the routers in a given FHRP group change (either based on routers being added to or removed from the FHRP group, or when a given router elects to withdraw from an FHRP group on the basis of link failures.
  • FIG. 8 illustrates a method 800 for n-way gateway load balancing in fabric path switching networks, according to certain embodiments of the present disclosure.
  • Method 800 assumes that an access switch has been configured according to the methods described above.
  • the method 800 begins at step 805 , where the access switch receives a frame from an attached host.
  • the access switch may connect a plurality of classical Ethernet (CE) hosts to an emulated switch itself constructed from multiple gateway routers.
  • CE classical Ethernet
  • the access switch identifies a destination address in the received frame. That is, the access switch identifies a MAC address in the destination address field of a MAC header in the received frame.
  • the access switch determines whether the destination address indicates whether received frame should be forwarded to the emulated switch. For example, using the network topology discussed above, the access switch 105 determines whether the received frame has a destination address of “VMAC.”
  • the access switch forwards the frame over a link which the destination address may be reached (step 820 ). Otherwise, at step 825 , if the destination address indicates a destination within the data center Ethernet or fabric path network (e.g., a destination address of the virtual MAC address assigned to an emulated switch) then the access switch wraps the frame with an outer MAC header (creating a MAC-in-MAC frame).
  • the other MAC frame header includes a source address of the access switch and a destination address of the emulated switch. Additionally, the access switch may select a link by evaluating a switch table which lists one or more links over which the emulated switch is reachable.
  • the access switch may select one in a manner to load balance the switch over each available link. As noted, the load balancing may distribute traffic to each link relatively equally, but could also favor one link over another (e.g., in the case where one link has a known greater capacity or bandwidth).
  • the access switch forwards the MAC-in-MAC frame over the selected link.
  • the example embodiments described above provide for n-way gateway load-balancing in a fabric path switching systems (FSS), such as a layer 2 multipath (L2MP) network in a large data center.
  • FSS fabric path switching systems
  • L2MP layer 2 multipath
  • an access switch in a L2MP network may learn an emulated switch IDs through FHRP “hello” messages sent by an “active” gateway router for each FHRP group. Further, the access switch may learn which links an emulated switch is reachable over by evaluating link state messages (e.g., IS-IS messages) sent by the gateway routers used to construct the emulated switch. Doing so allows the access switch to learn both the ID of an emulated switch and a set of links over which the emulated switch may be reached, thereby allowing the access switch to load balance traffic to the emulated switch.
  • link state messages e.g., IS-IS messages
  • aspects of the present disclosure may be implemented in hardware or software or in a combination of hardware and software.
  • One embodiment of the disclosure may be implemented as a program product for use with a computer system.
  • the program(s) of the program product define functions of the embodiments (including the methods described herein) and can be contained on a variety of computer-readable storage media.
  • Illustrative computer-readable storage media include, but are not limited to: (i) non-writable storage media (e.g., read-only memory devices within a computer such as CD-ROM disks readable by a CD-ROM drive, flash memory, ROM chips or any type of solid-state non-volatile semiconductor memory) on which information is permanently stored; and (ii) writable storage media (e.g., floppy disks within a diskette drive or hard-disk drive or any type of solid-state random-access semiconductor memory) on which alterable information is stored.
  • non-writable storage media e.g., read-only memory devices within a computer such as CD-ROM disks readable by a CD-ROM drive, flash memory, ROM chips or any type of solid-state non-volatile semiconductor memory
  • writable storage media e.g., floppy disks within a diskette drive or hard-disk drive or any type of solid-state random-access semiconductor memory

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Small-Scale Networks (AREA)
US13/331,626 2011-10-18 2011-12-20 Optimizations for N-way gateway load balancing in fabric path switching networks Expired - Fee Related US8717888B2 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
PCT/US2012/060792 WO2013059440A1 (en) 2011-10-18 2012-10-18 Fhrp optimizations for n-way gateway load balancing in fabric path switching networks
EP12783754.0A EP2769515B1 (de) 2011-10-18 2012-10-18 Fhrp-optimierungen für n-fachen gateway -lastausgleich in fps-netzen
CN201280051108.2A CN103891216B (zh) 2011-10-18 2012-10-18 结构路径交换网络中网关负载平衡的fhrp优化的方法和系统

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IN2986/DEL/2011 2011-10-18
IN2986DE2011 2011-10-18

Publications (2)

Publication Number Publication Date
US20130094357A1 US20130094357A1 (en) 2013-04-18
US8717888B2 true US8717888B2 (en) 2014-05-06

Family

ID=48085913

Family Applications (1)

Application Number Title Priority Date Filing Date
US13/331,626 Expired - Fee Related US8717888B2 (en) 2011-10-18 2011-12-20 Optimizations for N-way gateway load balancing in fabric path switching networks

Country Status (4)

Country Link
US (1) US8717888B2 (de)
EP (1) EP2769515B1 (de)
CN (1) CN103891216B (de)
WO (1) WO2013059440A1 (de)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140143591A1 (en) * 2012-11-19 2014-05-22 International Business Machines Corporation Performing Failover in a Redundancy Group
US9455948B2 (en) 2012-06-29 2016-09-27 Cisco Technology, Inc. Reducing proliferation of network-to-link-layer address resolution messages
US10892992B2 (en) 2016-07-01 2021-01-12 Hewlett Packard Enterprise Development Lp Load balancing
US12101252B2 (en) 2022-08-25 2024-09-24 Keysight Technologies, Inc. Methods, systems, and computer readable media for implementing routing path groups between emulated switches

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8681802B2 (en) 2011-08-15 2014-03-25 Cisco Technology, Inc. Proxy FHRP for anycast routing service
US8923296B2 (en) * 2012-02-23 2014-12-30 Big Switch Networks, Inc. System and methods for managing network packet forwarding with a controller
US9077562B2 (en) 2012-06-08 2015-07-07 Cisco Technology, Inc. System and method for layer-2 multicast multipathing
US9178837B2 (en) 2012-07-17 2015-11-03 Cisco Technology, Inc. System and method for layer-2 network routing
CN102769568B (zh) * 2012-07-19 2015-10-21 中兴通讯股份有限公司 一种基于虚拟交换集群的流量转发方法及系统
US9008095B2 (en) 2012-10-02 2015-04-14 Cisco Technology, Inc. System and method for hardware-based learning of internet protocol addresses in a network environment
US8948181B2 (en) * 2012-10-23 2015-02-03 Cisco Technology, Inc. System and method for optimizing next-hop table space in a dual-homed network environment
US9253140B2 (en) 2012-11-20 2016-02-02 Cisco Technology, Inc. System and method for optimizing within subnet communication in a network environment
CN103873368B (zh) * 2012-12-11 2017-09-15 新华三技术有限公司 以太网报文转发方法和接入设备
US9559962B2 (en) * 2013-01-22 2017-01-31 Brocade Communications Systems, Inc. Optimizing traffic flows via dynamic routing protocol modifications when using server virtualization with dynamic routing
US9088438B2 (en) 2013-01-29 2015-07-21 Cisco Technology, Inc. Using Ethernet ring protection switching with computer networks
CN104601346B (zh) 2013-10-30 2018-09-11 联想企业解决方案（新加坡）私人有限公司 管理交换机的网络连接的方法和装置
US9497104B2 (en) * 2014-01-14 2016-11-15 Cisco Technology, Inc. Dynamic update of routing metric for use in routing return traffic in FHRP environment
JP6211975B2 (ja) * 2014-03-27 2017-10-11 株式会社日立製作所 ネットワーク延伸システム、制御装置、およびネットワーク延伸方法
US9729498B2 (en) * 2014-05-21 2017-08-08 Alcatel Lucent Distributed address resolution protocol forwarding
CN104270280B (zh) * 2014-09-02 2017-11-07 烽火通信科技股份有限公司 在路由器上实现LSP ping和tracert的系统及方法
CN105813055B (zh) * 2014-12-30 2019-02-12 华为技术有限公司 基于ip的漫游方法和装置
US10033666B2 (en) * 2016-06-28 2018-07-24 Intel Corporation Techniques for virtual Ethernet switching of a multi-node fabric
CN107948086A (zh) * 2016-10-12 2018-04-20 北京金山云网络技术有限公司 一种数据包发送方法、装置及混合云网络系统
TWI813742B (zh) * 2018-08-23 2023-09-01 美商阿爾克斯股份有限公司 在網路路由環境中的非同步物件管理機制
CN113114569B (zh) * 2020-01-10 2023-06-02 北京京东尚科信息技术有限公司 数据流卸载的方法和装置
CN113254205B (zh) * 2021-05-24 2023-08-15 北京百度网讯科技有限公司 负载均衡系统、方法、装置、电子设备及存储介质

Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20010048661A1 (en) * 2000-05-24 2001-12-06 David Clear Method and apparatus for multi-protocol redundant router protocol support
US20020176363A1 (en) * 2001-05-08 2002-11-28 Sanja Durinovic-Johri Method for load balancing in routers of a network using overflow paths
US20070121617A1 (en) * 2005-11-29 2007-05-31 Cisco Technology, Inc. Extending sso for DHCP snooping to two box redundancy
US20080186968A1 (en) 2007-02-02 2008-08-07 Cisco Technology, Inc. Triple-tier anycast addressing
US20080225699A1 (en) * 2007-03-15 2008-09-18 Min-Kyu Lee Router and method of supporting nonstop packet forwarding on system redundant network
WO2009045130A1 (en) 2007-10-02 2009-04-09 Telefonaktiebolaget Lm Ericsson (Publ) Vrrp and learning bridge cpe
US20090257440A1 (en) 2006-12-22 2009-10-15 Huawei Technologies Co., Ltd. Method, system and router for communication between ip devices
US20100098082A1 (en) * 2008-10-16 2010-04-22 Rangaprasad Sampath Method for application broadcast forwarding for routers running redundancy protocols
US7716366B2 (en) * 2003-04-28 2010-05-11 Alcatel-Lucent Usa Inc. Enhancement of VRRP interface and router selection where an non-owner router is configured to respond to control and management messages addressed to an address associated with the virtual redundant router
US20110116509A1 (en) * 2009-11-16 2011-05-19 Moreno Victor M Method for the provision of gateway anycast virtual mac reachability in extended subnets
US8014290B2 (en) * 2003-10-17 2011-09-06 Hewlett-Packard Development Company, L.P. Traffic flow management through a multipath network
US8077709B2 (en) * 2007-09-19 2011-12-13 Cisco Technology, Inc. Redundancy at a virtual provider edge node that faces a tunneling protocol core network for virtual private local area network (LAN) service (VPLS)
US8166187B2 (en) * 2009-01-28 2012-04-24 Cisco Technology, Inc. Distributed IP gateway based on sharing a MAC address and IP address concurrently between a first network switching device and a second network switching device
US8315157B2 (en) * 2008-07-02 2012-11-20 Cisco Technology, Inc. Graceful removal and/or insertion of a network element

Patent Citations (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20010048661A1 (en) * 2000-05-24 2001-12-06 David Clear Method and apparatus for multi-protocol redundant router protocol support
US20020176363A1 (en) * 2001-05-08 2002-11-28 Sanja Durinovic-Johri Method for load balancing in routers of a network using overflow paths
US7716366B2 (en) * 2003-04-28 2010-05-11 Alcatel-Lucent Usa Inc. Enhancement of VRRP interface and router selection where an non-owner router is configured to respond to control and management messages addressed to an address associated with the virtual redundant router
US8014290B2 (en) * 2003-10-17 2011-09-06 Hewlett-Packard Development Company, L.P. Traffic flow management through a multipath network
US20070121617A1 (en) * 2005-11-29 2007-05-31 Cisco Technology, Inc. Extending sso for DHCP snooping to two box redundancy
US20090257440A1 (en) 2006-12-22 2009-10-15 Huawei Technologies Co., Ltd. Method, system and router for communication between ip devices
US8155131B2 (en) * 2006-12-22 2012-04-10 Huawei Technologies Co., Ltd. Method, system and router for communication between IP devices
US20080186968A1 (en) 2007-02-02 2008-08-07 Cisco Technology, Inc. Triple-tier anycast addressing
US20120300782A1 (en) * 2007-02-02 2012-11-29 Cisco Technology, Inc. Triple-tier anycast addressing
US8259720B2 (en) * 2007-02-02 2012-09-04 Cisco Technology, Inc. Triple-tier anycast addressing
US20080225699A1 (en) * 2007-03-15 2008-09-18 Min-Kyu Lee Router and method of supporting nonstop packet forwarding on system redundant network
US8077709B2 (en) * 2007-09-19 2011-12-13 Cisco Technology, Inc. Redundancy at a virtual provider edge node that faces a tunneling protocol core network for virtual private local area network (LAN) service (VPLS)
US20100220620A1 (en) * 2007-10-02 2010-09-02 Lars Timm VRRP and Learning Bridge CPE
WO2009045130A1 (en) 2007-10-02 2009-04-09 Telefonaktiebolaget Lm Ericsson (Publ) Vrrp and learning bridge cpe
US8315157B2 (en) * 2008-07-02 2012-11-20 Cisco Technology, Inc. Graceful removal and/or insertion of a network element
US20100098082A1 (en) * 2008-10-16 2010-04-22 Rangaprasad Sampath Method for application broadcast forwarding for routers running redundancy protocols
US8166187B2 (en) * 2009-01-28 2012-04-24 Cisco Technology, Inc. Distributed IP gateway based on sharing a MAC address and IP address concurrently between a first network switching device and a second network switching device
US20110116509A1 (en) * 2009-11-16 2011-05-19 Moreno Victor M Method for the provision of gateway anycast virtual mac reachability in extended subnets

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
International Search Report for application PCT/US2012/060792 dated Feb. 8, 2013.

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9455948B2 (en) 2012-06-29 2016-09-27 Cisco Technology, Inc. Reducing proliferation of network-to-link-layer address resolution messages
US20140143591A1 (en) * 2012-11-19 2014-05-22 International Business Machines Corporation Performing Failover in a Redundancy Group
US20140369183A1 (en) * 2012-11-19 2014-12-18 International Business Machines Corporation Performing Failover in a Redundancy Group
US9219641B2 (en) * 2012-11-19 2015-12-22 International Business Macines Corporation Performing failover in a redundancy group
US9219640B2 (en) * 2012-11-19 2015-12-22 International Business Machines Corporation Performing failover in a redundancy group
US10892992B2 (en) 2016-07-01 2021-01-12 Hewlett Packard Enterprise Development Lp Load balancing
US12101252B2 (en) 2022-08-25 2024-09-24 Keysight Technologies, Inc. Methods, systems, and computer readable media for implementing routing path groups between emulated switches

Also Published As

Publication number Publication date
US20130094357A1 (en) 2013-04-18
EP2769515A1 (de) 2014-08-27
CN103891216A (zh) 2014-06-25
EP2769515B1 (de) 2016-10-05
WO2013059440A1 (en) 2013-04-25
CN103891216B (zh) 2016-11-02

Similar Documents

Publication Publication Date Title
US8717888B2 (en) Optimizations for N-way gateway load balancing in fabric path switching networks
US10182496B2 (en) Spanning tree protocol optimization
US10333836B2 (en) Convergence for EVPN multi-homed networks
US7751329B2 (en) Providing an abstraction layer in a cluster switch that includes plural switches
US8565069B2 (en) Method of shrinking a data loss window in a packet network device
JP5873597B2 (ja) 仮想ファブリックリンク障害復旧のためのシステムおよび方法
JP5661929B2 (ja) マルチシャーシリンクアグリゲーションのためのシステムおよび方法
US7672253B2 (en) Border router with selective filtering of link state advertisements
US10075394B2 (en) Virtual link aggregations across multiple fabric switches
EP3378193A1 (de) Wahl und wiederwahl eines designated-forwarder (df) bei provider-edge(pe)-ausfall in einer vollaktiven redundanztopologie
US20130064102A1 (en) Virtual chassis system control protocols
US20190182202A1 (en) System and method for route optimization in a multichasiss link aggregation configuration
KR101691759B1 (ko) 가상 섀시 시스템 제어 프로토콜
US20170078150A1 (en) Ip-based interconnection of switches with a logical chassis
EP2502389A2 (de) Verfahren für virtuelle anycast-mac-erreichbarkeit eines gateway in erweiterten subnetzwerken
US12081458B2 (en) Efficient convergence in network events
KR101665276B1 (ko) 가상 섀시 시스템에서 패스 스루 모드를 위한 시스템 및 방법

Legal Events

Date Code Title Description
AS Assignment

Owner name: CISCO TECHNOLOGY, INC., CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:SANKAR, RAMKUMAR;GOWDA, NAVEEN PETHANDLAHALLY NARAYAN;REEL/FRAME:027420/0428

Effective date: 20111219

STCF Information on status: patent grant

Free format text: PATENTED CASE

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551)

Year of fee payment: 4

LAPS Lapse for failure to pay maintenance fees

Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

FEPP Fee payment procedure

Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

STCH Information on status: patent discontinuation

Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362

FP Lapsed due to failure to pay maintenance fee

Effective date: 20220506